← PicsartCONTENT HISTORY

Update to Picsart

Snapshot Nov 3, 2026 · 18:14 UTC · version 2.3.0

Collection source: downloaded plugin package.

WHAT CHANGED

First saved snapshot

No earlier snapshot is available to establish a change.

Save Picsart and check future changes in Following.

Up to 3 plugins free · no card required.

Compare saved observations

Download comparison JSON
Full technical diff · 0 changed fields
Full snapshot data
{
  "description": "Remix a reference reel, a talking head or a skit included - copy the reel's structure (beats and timing, shots, caption pattern, pace, sound bed) onto the user's own script or topic, never its words, faces, frames, audio or music; build it as slides, voice and video, and report what matched and what changed. Use for \"copy this reel's structure\" or \"remix this reel with my script\", from the video or a written breakdown. Not for rebuilding a reference's animation with the user's photos (recreate-video).",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 221
    }
  ],
  "name": "remix-reel-structure",
  "skill_md_contents": "---\nname: remix-reel-structure\ndescription: Remix a reference reel, a talking head or a skit included - copy the reel's structure (beats and timing, shots, caption pattern, pace, sound bed) onto the user's own script or topic, never its words, faces, frames, audio or music; build it as slides, voice and video, and report what matched and what changed. Use for \"copy this reel's structure\" or \"remix this reel with my script\", from the video or a written breakdown. Not for rebuilding a reference's animation with the user's photos (recreate-video).\nmetadata:\n  uses: reel-structure, footage-analysis, retention-hooks, film-cast-design, film-shot-generation, film-cutting, ai-presenter, voice-over-generation, spoken-captions, screen-recording-layouts, screen-recording-zoom, photo-slideshow, film-finishing, background-music, quality-review, scene-building-basics\n  title: Remix a reel's structure\n  short-description: Copy a reference reel's structure onto your own script\n  argument-hint: '[reference] [script.md or topic]'\n  starter-prompt: Copy this reel's structure onto my script, and show me what matched.\n---\n\n# Remix a reel's structure\n\nA reference reel lends its **structure**; the user's video keeps its own **content**. The\nstructure and length come from the reference, the channel's look from the user's house style,\nand the words, people and screens from the user's script. Nothing the viewer can see or hear of\nthe reference reaches the result.\n\nRead picsart-video-rules first. The method (the read, the sheet, the mapping and the report) is\nthe reel-structure skill's; this use case decides, asks and builds.\n\n| Stage | Gate | Skills |\n|---|---|---|\n| 1. Take in the reference, the script or topic, and the house style | ASK | — |\n| 2. Read the structure, or parse a breakdown for free | SPEND | reel-structure, footage-analysis |\n| 3. The structure sheet and the content rules | REVIEW | reel-structure |\n| 4. Map the script onto the beats, with the matched/changed report | ASK | reel-structure, retention-hooks |\n| 5. Route by format | — | below |\n| 6. `slides`: a still per beat, captions in the reference's pattern and the house style, no sound | REVIEW | film-shot-generation, spoken-captions |\n| 7. `voice`: locked voices, paced to the sheet's words per second | SPEND, REVIEW | voice-over-generation |\n| 8. `video`: the format's shots, cut on the beats, allowed spans kept | SPEND, REVIEW | the route's skills, film-cutting, quality-review |\n| 9. Finish: grade, the sound bed as the sheet says, loudness, cover | REVIEW | film-finishing, background-music |\n| 10. Final report (matched, changed, checked) and deliver | DELIVER | reel-structure |\n\n## Not recreate-video\n\nBoth start from a reference video, and share nothing else:\n\n| | recreate-video | remix-reel-structure |\n|---|---|---|\n| Copies | layout, timing and motion, measured frame by frame | beats, shot grammar, caption and pacing pattern |\n| Keeps from the reference | the arrangement, with the user's photos and text in its slots | nothing visible or audible |\n| Makes | an editable animated scene | new footage: generated, filmed or screen, cut to the structure |\n| Matches at | the pixel and the motion | the beat: its role, its length within 15 %, its shot type, its caption timing |\n\n**A talking head or a live-action reel comes here, never to recreate-video**: it has no layout\nto rebuild, only a structure to follow. An animated promo or app ad whose look the user wants with\ntheir own content goes to recreate-video.\n\n## The record\n\nKeep the project record as picsart-video-rules says, named `remix-reel-structure.json`. It holds\nthe reference (its URL or the breakdown's path), the house style's path, the script's path, the\nsheet, the mapping and the report, the route, each stage's approval, the locked cast and voices,\n`allowed-spans.json` once the screen clips are placed, and `openDecisions`. A later request\n(`slides`, `voice` or `video`, a fix) reads it and resumes at the next stage.\n\n## Content rules\n\nWritten here and checked by reel-structure's report at stages 4 and 10:\n\n- no line of the reference's speech in the script: no run of 4 or more of its words;\n- no still or clip of the reference as a generation input; its stills are for the user to look at;\n- no audio or music of the reference;\n- the caption copy is the user's own; only its pattern (place, size, timing) is copied.\n\nA failed rule is fixed, never waived: rewrite the line or change the input. A rule with nothing\nto check is reported as not checked, never as passed.\n\n## 1. Take it in\n\nASK (with the host's own choice tool where it offers one, such as request_user_input in Codex's Plan mode, so the user taps an option; a Picsart board the step names asks there instead; only where neither works, the options numbered in one message, the recommended one first; in unattended pace, never ask: write it to the project record's openDecisions with the answer you assumed, as picsart-video-rules says) for what is missing, in one question, with every file in one upload panel, as\npicsart-video-rules says (Asset URLs):\n\n- the reference: the video, or a written beat-by-beat breakdown of it;\n- the user's script (rows of time, picture, voice line and on-screen text), or a topic to write\n  one from;\n- their house style, if they keep one (frame, length, captions, sound, voices, look), and the\n  locked cast and voices they already have;\n- their own screen clips, when the reel shows a phone or a screen.\n\nTake a supplied script, cast and voices as fact; ask only what the next stage lacks. The house\nstyle's values win over the reference's look; the reference sets only the structure and length.\n\n**Exit:** a reference (video or breakdown), a script or topic, and the house style or \"none\" are\nrecorded.\n\n## 2. Read the structure\n\n**A written breakdown skips the read, and is free.** Parse it as reel-structure says.\n\n**A video is read with the tools** as reel-structure says. Quote what the read costs, then\nSPEND (quote it in one line, then get the yes as an ASK with \"Yes, go ahead\" / \"Yes, up to N credits for this task\" / \"No\": the host's own choice tool where it offers one, otherwise those options numbered in one message; the host's own confirmation before a tool runs is not this yes; unless a spend limit the user stated covers it: N credits or an explicit no-limit (\"as much as needed\"), never inferred from words like \"go ahead\". Within it, ask nothing: quote it in one line (\"About 8 credits; 12 of 20 left after this\"), call it, then say what it spent and what remains, keeping a running total that includes failed calls, and stop to ask before the first call that would cross it. A money amount is asked once to restate it in credits; in unattended pace, never ask: call only with a consent or a stated spend limit already recorded, otherwise write the quote to pendingSpends and stop, as picsart-video-rules says). Read the reference's speech only for its timing and pace, never to reuse it.\n\n**Exit:** a sheet with every beat timed and named, and the read's cost recorded.\n\n## 3. The structure sheet\n\nShow the sheet as a table, with one still per beat from the reference when it was read (for the\nuser only), and the content rules above. REVIEW it: the user corrects a beat, a time or a name,\nand the sheet is fixed before anything is mapped.\n\n**Exit:** the user approved the sheet as the structure to copy.\n\n## 4. Map the script\n\nWith a topic and no script, write the script to the sheet's beats first, one row per beat at the\nsheet's pace (words a second), with retention-hooks for the hook and the payoff; then map it.\n\nMap the script onto the beats and run the report as reel-structure says. Show the report: each\nreference beat matched, changed or dropped with its reason, each added row of the user's own, and\neach content rule. A failed rule is fixed in the script before this stage closes.\n\nWhen a beat is changed or dropped, ASK (with the host's own choice tool where it offers one, such as request_user_input in Codex's Plan mode, so the user taps an option; a Picsart board the step names asks there instead; only where neither works, the options numbered in one message, the recommended one first; in unattended pace, never ask: write it to the project record's openDecisions with the answer you assumed, as picsart-video-rules says) one beat at a time whether to rework the script toward\nthe reference or keep the change. Keep the others in `openDecisions`.\n\n**Exit:** the script and its mapping approved, every content rule passed or not checked with its\nreason, and the report saved in the record.\n\n## 5. Route by format\n\nThe sheet's format chooses where the shots come from. None of these is a new pipeline:\n\n| Reference format | Shots come from | Skills |\n|---|---|---|\n| POV skit, live-action story | generated takes with a locked cast | film-cast-design, film-shot-generation, film-cutting |\n| Talking head | a generated presenter on the locked voice | ai-presenter, voice-over-generation |\n| Screen tutorial | the user's own screen clips, placed as allowed spans | screen-recording-layouts, screen-recording-zoom |\n| List or countdown | a shot or still per item, with text cards | film-shot-generation or photo-slideshow, spoken-captions |\n| Reaction | the reacting person (generated) and the clip reacted to | as a talking head |\n\nA reaction's clip must be the user's own or licensed: ASK (with the host's own choice tool where it offers one, such as request_user_input in Codex's Plan mode, so the user taps an option; a Picsart board the step names asks there instead; only where neither works, the options numbered in one message, the recommended one first; in unattended pace, never ask: write it to the project record's openDecisions with the answer you assumed, as picsart-video-rules says) whether they hold the right to show\nit before it is placed. Never put the reference itself in as the reacted-to clip.\n\nRecord the route. A format the table does not name takes the nearest row, said in one line.\n\n## Build in stages: slides, voice, video\n\nWhen the user asks for one stage by name (`slides`, `voice` or `video`, as the `story` command\npasses it), build only that stage and stop at its REVIEW. Each stage opens only once the user\napproved the one before it. A stage asked for early says, in one line, which approval is missing,\nand offers to show that stage again. Without a named stage, the reel runs straight through.\n\n| Stage | What it builds | Paid |\n|---|---|---|\n| `slides` | one still per beat of the mapped script, held for the beat's time, with its on-screen text in the reference's caption pattern, and no sound | the pictures |\n| `voice` | each line spoken in the cast's locked voice, paced to the sheet's words per second, laid on the approved slides, the stills re-timed to the real lines | the voice lines |\n| `video` | the route's shots, cut on the beats, each take with its line's approved voice as reference audio where the model takes it | the takes |\n\n### 6. Slides\n\nDraft the stills as film-shot-generation's key frames say, a still per beat; SPEND (quote it in one line, then get the yes as an ASK with \"Yes, go ahead\" / \"Yes, up to N credits for this task\" / \"No\": the host's own choice tool where it offers one, otherwise those options numbered in one message; the host's own confirmation before a tool runs is not this yes; unless a spend limit the user stated covers it: N credits or an explicit no-limit (\"as much as needed\"), never inferred from words like \"go ahead\". Within it, ask nothing: quote it in one line (\"About 8 credits; 12 of 20 left after this\"), call it, then say what it spent and what remains, keeping a running total that includes failed calls, and stop to ask before the first call that would cross it. A money amount is asked once to restate it in credits; in unattended pace, never ask: call only with a consent or a stated spend limit already recorded, otherwise write the quote to pendingSpends and stop, as picsart-video-rules says) on the batch.\nBuild the slides as a scene, as scene-building-basics says, the text placed as spoken-captions says\nin the house style (font, plate, safe area). REVIEW it, titled \"<Reel> - slides\".\n\n### 7. Voice\n\nSPEND (quote it in one line, then get the yes as an ASK with \"Yes, go ahead\" / \"Yes, up to N credits for this task\" / \"No\": the host's own choice tool where it offers one, otherwise those options numbered in one message; the host's own confirmation before a tool runs is not this yes; unless a spend limit the user stated covers it: N credits or an explicit no-limit (\"as much as needed\"), never inferred from words like \"go ahead\". Within it, ask nothing: quote it in one line (\"About 8 credits; 12 of 20 left after this\"), call it, then say what it spent and what remains, keeping a running total that includes failed calls, and stop to ask before the first call that would cross it. A money amount is asked once to restate it in credits; in unattended pace, never ask: call only with a consent or a stated spend limit already recorded, otherwise write the quote to pendingSpends and stop, as picsart-video-rules says) on the lines as one batch, voiced as voice-over-generation says, in the locked voices.\nA line's real length moves its still, never the words. REVIEW the voiced slides.\n\n### 8. Video\n\nBuild the route's shots as its skills say, every paid call quoted first under SPEND (quote it in one line, then get the yes as an ASK with \"Yes, go ahead\" / \"Yes, up to N credits for this task\" / \"No\": the host's own choice tool where it offers one, otherwise those options numbered in one message; the host's own confirmation before a tool runs is not this yes; unless a spend limit the user stated covers it: N credits or an explicit no-limit (\"as much as needed\"), never inferred from words like \"go ahead\". Within it, ask nothing: quote it in one line (\"About 8 credits; 12 of 20 left after this\"), call it, then say what it spent and what remains, keeping a running total that includes failed calls, and stop to ask before the first call that would cross it. A money amount is asked once to restate it in credits; in unattended pace, never ask: call only with a consent or a stated spend limit already recorded, otherwise write the quote to pendingSpends and stop, as picsart-video-rules says). Cut them on\nthe beats as film-cutting says. Place the user's screen clips first, write `allowed-spans.json` from\nwhere each lands, and never cut, re-time or zoom inside one. Check the cut with quality-review, then\nREVIEW it.\n\n### 9. Finish\n\nAs film-finishing says: the grade from the house style, loudness, and the cover. The sound is what\nthe sheet's bed says: voice only stays voice only; a bed is drafted with background-music, never the\nreference's music. REVIEW the finished reel.\n\n## 10. Report and deliver\n\nRun the report again on what was built, as reel-structure says: each beat matched, changed or\ndropped, and each content rule, with the media list (the reference's URLs, every generation\ninput, the audio in the cut), so pictures and sound are checked, not assumed. Say which lines are\nmeasured and which only follow the plan.\n\nDELIVER the master as picsart-video-rules says, with the report beside it.\n\n**Exit:** the master delivered, the report saved in the record, every content rule passed.\n\n## When something fails\n\nA failed read call is retried once, then the user is offered the free route: a written breakdown.\nOut of credits: no retry; say which stage paused and keep the record. A content rule that fails\nafter the build sends the reel back to the stage that put the reference's content in.\n"
}

SHA-256 of public snapshot: 022573b608f17f6399cf7a21709864e0839bccab64281b2a0899816e4f9bd6f8